Multi-PGS enhances polygenic prediction by combining 937 polygenic scores

Evaluation of polygenic scoring methods in five biobanks shows larger variation between biobanks than methods and finds benefits of ensemble learning

Methods of estimating polygenic scores (PGSs) from genome-wide association studies are increasingly utilized. However, independent method evaluation is lacking, and method comparisons are often limited. Here, we evaluate polygenic scores derived via seven methods in five biobank studies (totaling about 1.2 million participants) across 16 diseases and quantitative traits, building on a reference-standardized framework. We conducted meta-analyses to quantify the effects of method choice, hyperparameter tuning, method ensembling, and the target biobank on PGS performance. We found that no single method consistently outperformed all others. PGS effect sizes were more variable between biobanks than between methods within biobanks when methods were well tuned. Differences between methods were largest for the two investigated autoimmune diseases, seropositive rheumatoid arthritis and type 1 diabetes. For most methods, cross-validation was more reliable for tuning hyperparameters than automatic tuning (without the use of target data). For a given target phenotype, elastic net models combining PGS across methods (ensemble PGS) tuned in the UK Biobank provided consistent, high, and cross-biobank transferable performance, increasing PGS effect sizes (β coefficients) by a median of 5.0% relative to LDpred2 and MegaPRS (the two best-performing single methods when tuned with cross-validation). Our interactively browsable online-results and open-source workflow prspipe provide a rich resource and reference for the analysis of polygenic scoring methods across biobanks.

Using Machine Learning to Evaluate the Value of Genetic Liabilities in the Classification of Hypertension within the UK Biobank

Background and Objective: Hypertension increases the risk of cardiovascular diseases (CVD) such as stroke, heart attack, heart failure, and kidney disease, contributing to global disease burden and premature mortality. Previous studies have utilized statistical and machine learning techniques to develop hypertension prediction models. Only a few have included genetic liabilities and evaluated their predictive values. This study aimed to develop an effective hypertension classification model and investigate the potential influence of genetic liability for multiple risk factors linked to CVD on hypertension risk using the random forest and the neural network. Materials and Methods: The study involved 244,718 European participants, who were divided into training and testing sets. Genetic liabilities were constructed using genetic variants associated with CVD risk factors obtained from genome-wide association studies (GWAS). Various combinations of machine learning models before and after feature selection were tested to develop the best classification model. The models were evaluated using area under the curve (AUC), calibration, and net reclassification improvement in the testing set. Results: The models without genetic liabilities achieved AUCs of 0.70 and 0.72 using the random forest and the neural network methods, respectively. Adding genetic liabilities improved the AUC for the random forest but not for the neural network. The best classification model was achieved when feature selection and classification were performed using random forest (AUC = 0.71, Spiegelhalter z score = 0.10, p-value = 0.92, calibration slope = 0.99). This model included genetic liabilities for total cholesterol and low-density lipoprotein (LDL). Conclusions: The study highlighted that incorporating genetic liabilities for lipids in a machine learning model may provide incremental value for hypertension classification beyond baseline characteristics.

MUSSEL: Enhanced Bayesian polygenic risk prediction leveraging information across multiple ancestry groups

Polygenic risk scores (PRSs) are now showing promising predictive performance on a wide variety of complex traits and diseases, but there exists a substantial performance gap across populations. We propose MUSSEL, a method for ancestry-specific polygenic prediction that borrows information in summary statistics from genome-wide association studies (GWASs) across multiple ancestry groups via Bayesian hierarchical modeling and ensemble learning. In our simulation studies and data analyses across four distinct studies, totaling 5.7 million participants with a substantial ancestral diversity, MUSSEL shows promising performance compared to alternatives. For example, MUSSEL has an average gain in prediction R2 across 11 continuous traits of 40.2% and 49.3% compared to PRS-CSx and CT-SLEB, respectively, in the African ancestry population. The best-performing method, however, varies by GWAS sample size, target ancestry, trait architecture, and linkage disequilibrium reference samples; thus, ultimately a combination of methods may be needed to generate the most robust PRSs across diverse populations.

Integrative polygenic risk score improves the prediction accuracy of complex traits and diseases

Polygenic risk scores (PRSs) are an emerging tool to predict the clinical phenotypes and outcomes of individuals. We propose PRSmix, a framework that leverages the PRS corpus of a target trait to improve prediction accuracy, and PRSmix+, which incorporates genetically correlated traits to better capture the human genetic architecture for 47 and 32 diseases/traits in European and South Asian ancestries, respectively. PRSmix demonstrated a mean prediction accuracy improvement of 1.20-fold (95% confidence interval [CI], [1.10; 1.3]; p = 9.17 × 10-5) and 1.19-fold (95% CI, [1.11; 1.27]; p = 1.92 × 10-6), and PRSmix+ improved the prediction accuracy by 1.72-fold (95% CI, [1.40; 2.04]; p = 7.58 × 10-6) and 1.42-fold (95% CI, [1.25; 1.59]; p = 8.01 × 10-7) in European and South Asian ancestries, respectively. Compared to the previously cross-trait-combination methods with scores from pre-defined correlated traits, we demonstrated that our method improved prediction accuracy for coronary artery disease up to 3.27-fold (95% CI, [2.1; 4.44]; p value after false discovery rate (FDR) correction = 2.6 × 10-4). Our method provides a comprehensive framework to benchmark and leverage the combined power of PRS for maximal performance in a desired target population.

Machine learning and phylogenetic analysis allow for predicting antibiotic resistance in M. tuberculosis

BACKGROUND

Antimicrobial resistance (AMR) poses a significant global health threat, and an accurate prediction of bacterial resistance patterns is critical for effective treatment and control strategies. In recent years, machine learning (ML) approaches have emerged as powerful tools for analyzing large-scale bacterial AMR data. However, ML methods often ignore evolutionary relationships among bacterial strains, which can greatly impact performance of the ML methods, especially if resistance-associated features are attempted to be detected. Genome-wide association studies (GWAS) methods like linear mixed models accounts for the evolutionary relationships in bacteria, but they uncover only highly significant variants which have already been reported in literature.

RESULTS

In this work, we introduce a novel phylogeny-related parallelism score (PRPS), which measures whether a certain feature is correlated with the population structure of a set of samples. We demonstrate that PRPS can be used, in combination with SVM- and random forest-based models, to reduce the number of features in the analysis, while simultaneously increasing models' performance. We applied our pipeline to publicly available AMR data from PATRIC database for Mycobacterium tuberculosis against six common antibiotics.

CONCLUSIONS

Using our pipeline, we re-discovered known resistance-associated mutations as well as new candidate mutations which can be related to resistance and not previously reported in the literature. We demonstrated that taking into account phylogenetic relationships not only improves the model performance, but also yields more biologically relevant predicted most contributing resistance markers.

Digenic Analysis Finds Highly Interactive Genetic Variants Underlying Polygenic Traits

We briefly review our recently published approach to mining digenic genotype patterns, which consist of two genotypes each originating in a different DNA variant. We do this for a genetic case-control study by evaluating all possible pairs of genotypes, distributing the workload over numerous CPUs (threads) in a high-performance computing environment and apply our methods to two known datasets, age-related macular degeneration (AMD) and Parkinson Disease (PD). Based on a list of (e.g., 100,000) genotype pairs with largest genotype pair frequency differences between cases and controls, we determine the numberN u of unique variants occurring in this list. For each unique variant, we find the number of genotype pairs it participates in, which identifies a set of variants "connected" with the given unique variant. Among the total of variants "connected" with all unique variants, only a subset of variants is unique. The ratio of all connected variants divided by that subset of variants is a measure for the overall density or connectedness of variants interacting with each other. We find that variants for the AMD data are much more interconnected than those for PD, at least based on the 100,000 genotype pairs with largest chi-square we investigated. Further, for each of theN u unique variants, we use the number of variants connected with it as a test statistic, weighted by the inverse of the rank at which the unique variant first occurred in the original list of genotype patterns. This weighing scheme ties the number of connections to the genetics of the trait and allows us to obtain, for each of theN u unique variants, an empirical significance level by permuting ranks. We find 12 and 8 significant, highly connected variants for AMD and PD, respectively, some of which have previously been identified by other machine learning methods, thus providing credence to our approach. Among the 100,000 genotype pairs investigated for each of AMD and PD, significant variants showed connections with up to 7,093 and 3,777 other variants, respectively. Our approach has been implemented in a freely available piece of software, the Digenic Network Test. Thus, our statistical genetics method can provide important information on the genetic architecture of polygenic traits.